Marco-Christiani/zigrad

A deep learning framework built on an autograd engine with high level abstractions and low level control.

46
/ 100
Emerging

Zigrad helps AI researchers and engineers quickly transition their deep learning experiments into high-performance training systems. It takes early-stage deep learning model designs, often created with PyTorch-like interfaces, and outputs highly optimized, fast-running models ready for large-scale training. This tool is for AI practitioners and MLOps engineers who need to bridge the gap between rapid prototyping and production-ready performance.

186 stars.

Use this if you are developing deep learning models and need to optimize their performance for large-scale training without rebuilding your entire workflow or switching frameworks.

Not ideal if you need a comprehensive library of pre-built, optimized deep learning layers like convolutions or pooling, as these are still under development.

deep-learning-research mlops model-optimization ai-development high-performance-computing
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

186

Forks

10

Language

Zig

License

LGPL-3.0

Last pushed

Mar 11, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Marco-Christiani/zigrad"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.